Discussion List Archives

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Background to our discussion

> Please send any comments to the 'reply to' address before
> November 15.

  Might even make it on time for once.

> 2. Background

  I wish to caution that there are problems with reducing the
description of a crystal pattern to that of a set of points. The
symmetry of a set of points is not the same as the symmetry of a set of
atoms occupying those points. e.g. take the sodium chloride structure
and replace each atom by a point. NaCl structure occurs in a cubic F
space group, whereas the related structure with atoms replaced by points
is primitive. Likewise David talks about a point which is the centre of
mass of the molecule. You can not find this just from the coordinates of
the points representing the centre of mass of each atom - you need the
mass as well. In my mind a crystal structure looks more naturally like a
set of atoms (each with a position given by the coordinates of its
centre of mass and other properties like element type) rather than a set
of points (of known coordinates which acquire elemental properties).

> to consider how to identify molecules and other chemical entities

  Before finding a way to identify these entities we need to be sure we
agree on what we consider to be a molecule or 'other chemical entities'.
I feel that it is not our problem to define these. We should use
definitions acceptable and accepted by chemists. I hope that IUPAC has
something to meet these needs. This is critical for the description of
molecular symmetry. E.g. if polyethylene is not a molecule, we can stick
with point groups to describe the symmetry properties of molecules; if
polyethylene is a molecule, I recommend you all go out and immediately
buy a copy of International Tables for Crystallography Volume E.

> Atoms, .... are ... chemical rather than crystallographic entities.

  You have already seen that I don't quite agree.

> 4. Chemical entities and properties we may wish to define:

  Of course you all know that chemically-equivalent molecules may not be
crystallography-equivalent. However we have to face up to the horrid
truth that crystallography-equivalent molecules may not be

> 4.1.1 Atoms:

   Add name (label) and chirality.

> 4.1.3 Molecules:

   Add name, quadruple moment.
   The optical rotation is a property of the compound or substance not
of the molecule.

   I also have a fancy that sets or collections of molecules need a

   In all of this, racemates are going to be a nightmare.
> In all the above cases the atoms used in defining the chemical
> properties must be linked to the crystallographic positions they
> occupy.

  The current 'ORTEP'-like code for doing translations in
symmetry-related atoms needs a rethink. -5 to +4 is not enough. If the
molecule is allowed to have translational symmetry, some other technique
is necessary.

> 5.4
> It is the only item in the
> chemical category that is not related to the crystal.

  I'm not quite sure what you are driving at in this sentence. In my
view something like 'chemical_absolute_configuration' is definitely only
related to a particular molecule and has nothing to do at all with the
crystal, whereas '_chemical_optical_rotation' is a property of the whole
substance (of the crystals taken into solution) and is not the property
of an individual molecule.

coreCIFchem mailing list

[Send comment to list secretary]
[Reply to list (subscribers only)]